The present invention relates generally to biodegradable polymers. More particularly, it concerns methods for the bioproduction of novel hydroxy-terminated polyhydroxyalkanoate (PHA) polymer compositions, and their subsequent use in production of novel copolyesters, polyester carbonates, polyester ethers, polyester urethanes, polyester amides, polyester acetals and other elastomeric, thermoplastic and thermoset polymers and copolymers.
There has been considerable interest in recent years in the use of biodegradable polymers to address concerns over plastic waste accumulation. The potential worldwide market for biodegradable polymers is enormous (&gt;10B lbs/yr). Some of the markets and applications most amenable to the use of such biopolymers range from single use applications, which include packaging, personal hygiene, garbage bags, and others where the biopolymers become soiled and are ideally suited for biodegradation through composting, to markets and applications in which the biopolymers can be recovered as clean materials, such as garment bags, shopping bags, grocery bags, etc. and are suitable for recycling, as well as composting, or biodegradation in landfills.
PHA biopolymers are thermoplastic polyesters produced by numerous microorganisms in response to nutrient limitation. The commercial potential for PHA's spans many industries, and is derived primarily from certain advantageous properties which distinguish PHA polymers from petrochemical-derived polymers, namely excellent biodegradability and natural renewability.
Widespread use and acceptance of PHA's, however, has been hindered by certain undesirable chemical and physical properties of these polymers. For example, PHA's are among the most thermosensitive of all commercially available polymers. As such, the rate of polymer degradation, as measured by a decrease in molecular weight, increases sharply with increasing temperatures in the range typically required for conventional melt-processing of PHA's into end-products such as films, coatings, fibers etc. An additional limitation of the potential utility of PHA polymers relates to the observation that some polymer characteristics, for example ductility, elongation, impact resistance, and flexibility, diminish over time. This rapid "aging" of certain PHA-derived products is unacceptable for many applications. Thus, the success of PHA as a viable alternative to both petrochemical-derived polymers and to non-PHA biodegradable polymers, will depend upon novel approaches to overcome the unique difficulties associated with PHA polymers and with products derived therefrom.
One approach which has the potential to provide new classes of PHA-containing polymers having unique and improved properties, is based on graft, random and block polymers and copolymers. In generating such polymers and copolymers, it is possible to vary the nature, length and mass fraction of the different polymer constituents which are present. In doing so, the morphology of the polymer, and therefore the resulting properties, may be manipulated to meet the requirements of a given application.
The production of copolymers with PHA is limited, however, by the dissimilar ends of a PHA polymer chain (i.e. a carboxy group and a hydroxy group, respectively, on the ends of each polymer molecule). In order for PHA to be useful in the production of copolymers, it is desired that the ends of a polymer molecule chain possess the same chemical groups, and that those groups be capable of forming covalent bonds with the ends of other polymer molecules either by direct reaction or by use of a coupling agent (e.g. a diisocyanate). The reactive end groups of hydroxy-terminated PHA, for example, would be well suited for the preparation of high MW block copolymers by various known chain extension approaches.
Synthetic PHA hydroxy-termination has been reported. Hirt et al. (Macromol. Chem. Phys. 197, 1609-1614 (1996)) describe a method for the preparation of HO-terminated PHB and PHB/HV by a transesterification procedure using ethylene glycol and commercial grade PHB and PHBV (Biopol) in the presence of a catalyst. When catalysts such as H.sub.3 PO.sub.4, ethylene glycolate, or tripropyl amine were used, some diol was formed but the major products were oligomers with carboxylic acid end groups and olefinic end groups, even when a tenfold excess of ethylene glycol was used. When dibutyltin-dilaurate was used as catalyst with a 10-fold excess of ethylene glycol, end-group hydroxyl incorporation was obtained. The oligomer had a hydroxyl end-group content of .about.97%, but the molecular weight (Mn) of the hydroxy-terminated PHA obtained by Hirt et. al. was only about 2,300. The ratio of the weight- to number-average molecular weights (Mw/Mn) of their oligomers was .about.2.
When one prepares hydroxy-terminated PHA by reaction of PHA with EG in the presence of a catalyst, a steady decline in MW to less than about 3,000 is observed as the content of hydroxyl end groups increases from about 50% to greater than 80%. Experiments performed in our laboratory confirm this effect. (the results of which are presented in Example 1 and in FIG. 1).
For many applications, higher MW hydroxy-terminated PHA would be preferred for use in the preparation of copolymers. In addition, it would be advantageous to have the ability to produce hydroxy-terminated PHA by a simple and inexpensive method that is easily amenable to large scale polymer production.
Shi et al (Macromolecules 29 10-17 (1996)) reported in vivo formation of a hybrid natural-synthetic di-block copolymer. They describe a method to prepare PHA-polyethylene glycol (PEG) di-block copolymers where the carboxylate terminus of PHA chains are covalently linked by an ester bond to PEG chain segments. Polymer production was carried out by cultivation of A. eutrophus in a nitrogen free medium containing 4-HBA and 4% w/v PEG (Mn.about.200). The unfractionated product was complex in that it was composed of at least three different component polymers of different repeat unit composition. The product was separated into an acetone insoluble fraction (43% w/w, Mn.about.130 k, Mw/Mn.about.3.42) and an acetone soluble fraction (57% w/w, Mn.about.37.4 k, Mw/Mn.about.2.52). The mole fraction content of 3HB, 3HV, 4HB, and PEG in the acetone insoluble and acetone soluble fractions were 95, 2, 3, 0.1 and 13, 2, 84, 1.6, respectively. Thus the PEG segments were found primarily in the acetone soluble fraction. It is important to note that the acetone soluble fraction is high in 4HB while the acetone insoluble fraction is high in 3HB. This process to make the di-block copolymer is not very selective.
Shi et. al. (Macromolecules 29 7753-8 (1996)) later reported the use of PEG's to regulate MW during A. eutrophus cultivations. They used PEG's that ranged in MW from is PEG-106 (diethylene glycol) up to PEG-10,000 (MW.about.10 k) and reported that PEG-106 was the most effective in that only 0.25% was required to reduce Mn of the PHA by 74%. By adding 10% PEG-106 to the medium, the decrease in Mn was from 455 k to 19.4 k. They also reported that the use of monomethoxy ether CH3O-PEG-OH-350 and PEG-300 resulted in almost identical molecular weight reductions. However, the dimethoxy ether of tetraethylene glycol was not an effective agent for molecular weight reduction. Thus, the chain end functionality useful in the work was a hydroxyl group. Shi et al. also reported that their PHA products did not contain PEG terminal groups. They concluded that molecular weight reduction was not due to chain termination reactions by PEG, but likely due to interaction between PEG and the PHA production system, leading to an increased rate of chain termination by water relative to chain propagation reactions.
We have now discovered that by using simple, monomeric aliphatic diols or aliphatic polyols, that clean and selective reactions can be obtained by in vivo formation of PHA-diols at concentrations of aliphatic diols or aliphatic polyols in the culture medium that are not toxic to the PHA-producing microorganisms.